
The Federal Reserve Bank of Atlanta lowered its GDPNow estimate of U.S. economic growth in the fourth quarter of 2026 to 3.2% on November 18, down from 3.4% a day earlier. The model’s latest reading represents real gross domestic product growth at a seasonally adjusted annual rate, not an official tally of output for the quarter.
The two-tenths-of-a-percentage-point reduction leaves the estimate below the 3.5% reading recorded on November 9. Even after the downgrade, GDPNow points to continued economic expansion if the forecast is borne out. It does not suggest that output is contracting, and it is not a Federal Reserve interest-rate decision or an official policy forecast.
GDPNow gives back earlier gains as fresh data arrive
The fourth-quarter model began with a 3.2% estimate on October 29, then climbed to 3.7% on November 2. By November 4, that figure had dropped to 3.4%, before edging back up to 3.5% on November 9. The latest move brings the nowcast back to its starting level, illustrating how much a reading can change as more of the quarter’s economic data become available.
Tuesday’s revision from 3.5% to 3.4% came with an explanation from the Atlanta Fed. Its November 17 commentary showed that the model’s estimate for fourth-quarter real consumer spending growth had moved down from 2.6% to 2.5%, while the projection for real gross private domestic investment growth declined from 10.7% to 10.2%. An upward adjustment to government-spending assumptions partly countered those changes.
Those component figures describe the November 17 revision, not the subsequent decline to 3.2%. The Atlanta Fed’s research-data feed showed the new headline estimate on November 18, but its publicly available commentary had not yet provided a detailed breakdown of what accounted for that day’s change. Assigning the additional reduction to a particular report would therefore go beyond the published explanation.
One of the day’s scheduled inputs was the Census Bureau’s October residential construction report. Housing starts were estimated at an annual rate of 1.309 million units, up 9.9% from the revised September estimate, while building permits declined 7.4% to 1.295 million units. The starts increase was subject to a wide statistical margin of error. These mixed construction figures give context for the housing sector but do not, by themselves, establish why GDPNow moved lower.
Other figures incorporated around this stage of the quarter include retail sales, business inventories, import and export prices, and industrial production. They matter for different parts of GDP, from household purchases to business investment and trade. Because the model aggregates numerous components, a stronger reading in one area need not produce an increase in the overall growth estimate.
Why the 3.2% reading is not an official Fed forecast
The Atlanta Fed describes GDPNow as a running, data-driven estimate of the growth rate the Bureau of Economic Analysis will eventually report. It is designed to mirror parts of the government’s GDP accounting process, using available monthly source statistics and statistical projections for information that has not yet been released.
The model builds an overall estimate from forecasts of 13 GDP components rather than asking economists to choose a growth rate. Its calculations are updated after important economic releases, without discretionary adjustments to make the number fit a preferred outlook. As a result, a revision may reflect new information, revisions to earlier data or the way those inputs affect several parts of the economy.
There is also a difference between an annualized growth rate and growth over a single three-month period. A 3.2% annualized estimate expresses the quarter’s pace as though it continued for a full year; it does not mean U.S. production rose 3.2% between October and December alone. The distinction matters when comparing the nowcast with monthly indicators or year-over-year measures.
GDPNow is not an official forecast from the Atlanta Fed president, the Federal Reserve System or the Federal Open Market Committee. A lower reading can inform discussion of economic momentum, but it cannot establish how policymakers will vote on interest rates. Monetary-policy decisions consider a broader body of evidence, and the bank explicitly warns that forecasting errors can remain substantial even near the first official GDP release.
The estimate will keep changing as the fourth quarter progresses. Some source figures are available relatively quickly, while others arrive with a lag or are later revised. The model fills those gaps statistically until the incoming information replaces its assumptions, which is why a reading from mid-November should not be treated as a final result for December.
How the estimate compares with the latest measured GDP growth
The most recent official quarterly GDP report provides a benchmark, although it measures a different period. In its advance estimate for the third quarter, the Bureau of Economic Analysis reported real GDP growth of 3.4% at an annual rate, following 2.2% growth in the second quarter. The third-quarter figure was released on October 29 and remains subject to revisions.
According to the BEA, the third-quarter expansion reflected increased investment, consumer spending and government spending, partly offset by a decrease in exports. Imports also rose and are subtracted in calculating GDP. Those details show why the headline growth rate can move even when some underlying parts of demand remain strong.
A fourth-quarter nowcast of 3.2%, compared with the third quarter’s initial 3.4% figure, would imply a modest easing in annualized growth if both readings ultimately held. That is a comparison between a preliminary model projection and an already published government estimate, however, not evidence that the fourth quarter has finished with weaker growth. The official third-quarter number could also change as more complete source data are incorporated.
For a clearer picture of the economy’s direction, the forthcoming spending, investment and trade releases will be more informative than any single update to the top-line model. The BEA has scheduled its second estimate of third-quarter GDP for November 25, the same date on which the Atlanta Fed’s GDPNow calendar lists another fourth-quarter update tied to additional economic reports.
Latest News
View all news- TJX Reports 4% Comparable-Sales Growth and $1.34 in Adjusted Quarterly Earnings Per Share
- McGraw Hill Sets 5%–8% Long-Term Revenue Growth Target, Outlines Cash Flow Plan
- Infinity Natural Resources Signs Reorganization Agreement, Targets December 1 Closing
- U.S. Outdoor Recreation Economy Reaches $704.2 Billion, New Federal Data Show
- U.S. Pending Home Sales Fall 4.7% in October, Marking Second Straight Month of Record-Low Contracts